Skip to content
Xweather home

Hail intelligence from forecast to validation

The industry's only hail forecast API, validated historical data, and advanced mapping layers. Predict threats at the asset level, alert before impact, and verify what happened after the storm.

Best in the world

  • 99.2 %

    of U.S. severe hailstorms predicted in H1 2026

  • 53 min

    mean lead time

  • ≥ 15 min

    warning for 95.9% of severe hailstorms

  • Validated on 3,659

    severe hail reports from the U.S.

A 58 million dollar problem

Hail accounts for over 50% of solar facility claim costs

For utility-scale solar, hail damage represents less than 2% of filed claims but over 50% of total claim costs, with the average solar hail claim reaching $58 million (AXIS Capital, 2024). Insurers have responded with premium increases of up to 400% in high-risk regions, sub-limits on hail coverage, and in some cases refusals to insure at all.

The risk extends beyond solar. Any organization with outdoor infrastructure, from warehouses and vehicle fleets to airports and construction sites, faces hail exposure. Standard severe weather warnings cover wide areas but cannot tell you exactly where hail will fall or how long it will last.

That gap between a general warning and precise, site-level hail intelligence is where damage, downtime, and disputed claims happen.

The cost of imprecise hail warnings

Unpredictable strike zones

Severe storm warnings cover hundreds of square kilometers. Hail is localized within a few kilometers. Without site-level forecasts, assets remain exposed to threats that regional warnings cannot resolve.

Narrow reaction windows

Hail-producing storms can intensify in minutes. Stowing solar panels, pausing outdoor operations, or moving vehicles requires lead time that generic storm warnings rarely provide.

Disputed claims, unclear exposure

After a hailstorm, insurers and asset owners often disagree on what happened and where. Without validated, location-specific sensor data on hail severity and duration, claims take longer to settle, and exposure is harder to price.

The only hail threat forecast powered by lightning intelligence

Most hail forecasts rely on radar, satellite, and standard atmospheric models. Xweather adds proprietary lightning data that others don't have.

Lightning as a core signal of storm development

Xweather is the only provider that integrates lightning data directly into hail threat forecasting, combining it with radar, satellite, and advanced modeling to detect storm intensity and hail potential earlier and with greater precision.

How the hail forecast model works

Our forecast model combines lightning data from the Xweather Lightning Network with radar reflectivity and numerical weather models to forecast hail threats up to 60 minutes in advance. 

A fundamentally different approach to forecasting hail risk

With this approach, we identify high-risk conditions that conventional methods miss. The result is a more accurate, more reliable hail threat forecast, validated by performance data and trusted in operational environments where decisions carry real cost.

Where hail intelligence matters most

Solar and energy

Protect solar assets before hail hits

Hail forecast alerts delivered up to 60 minutes before impact, with the probability of severe hail, time of arrival, storm approach direction, and the recommended stow direction.

Stow protocols have been shown to materially reduce damage exposure and lower claims frequency (AXIS Capital).

After the storm, the historical archive confirms hail size and severity at your exact location, supporting damage assessments, warranty claims, and insurance documentation.

Insurance and risk

Validate claims with verified hail records

Access location-specific historical data on hail size, frequency, severity, duration, and intensity for any covered location.

Use validated records to assess policy exposure, price hail risk accurately, and resolve claims faster.

Archive data is available from January 2024 (U.S.) and February 2024 (Canada, Japan, Europe, Australia) to present day.

Wilted corn plants lying flat on muddy ground, damaged by a a hail storm.
forecast icon white

Validation study: Xweather predicts 99.2% of severe hailstorms

Tested against 3,659 severe hail reports from the U.S., Xweather outperformed National Weather Service warnings and the radar-based method behind most commercial services in both accuracy and lead time.

A smooth gradient ranging from dark to lighter shades of blue, creating a serene and calming abstract background.
Rows of solar panels with storm clouds approaching in the background
  • Outperforms radar by 15.9 percentage points

    Xweather predicted 99.2% of the severe hail events in the study. Radar detected 83.3%, missing roughly one in six storms outright.

  • 53 minutes mean lead time

    Xweather provided the highest mean lead time at 53 minutes, against 50 minutes for radar and 30 for NWS.

  • At least 15 minutes warning for 95.9% of events

    Xweather provided at least 15 minutes of warning for 95.9% of the reported events, compared with 72.2% for radar.

Create hail alerts with Xweather Protect

01
Connect

Add your assets in Xweather Protect, configure the Alert API add-on, and define an alerting buffer zone around your location.

02
Monitor

Monitor developing hailstorms with forecast polygons showing impacted areas up to 60 minutes ahead.

03
Alert

Alerts and webhook notifications fire when hail is forecast near your assets.

04
Respond

Stow panels, pause operations, or activate protection protocols using hail size, probability, and storm track data.

05
Verify

Access the historical archive to confirm hail size, severity, and duration at each location for claims and reporting.

A storm approaches a rural landscape with dramatic, dark clouds covering the sky, casting shadows over fields and a small garden.

Data to solve the hail problem

Weather API misty gray

Hail threat nowcasts via API

Xweather's hail threats endpoint delivers forecasts up to 60 minutes ahead in 10-minute intervals. The Alert API activates alerts on the probability of severe hail ( 2.54 cm diameter or larger) within a configurable buffer distance from your asset.

hail misty gray

Historical hail data archive

Validated historical hail data dating back to early 2024. Records include hail size, frequency, severity, duration, and intensity for any covered location. Built for insurance, claims validation, exposure modeling, and meteorological research.

MapsGL misty gray

Real-time hail mapping with MapsGL

Eight dedicated hail map layers, updated every 2 minutes, display forecast threat areas, storm tracks, hail size, and severe hail probability. Available via JavaScript, iOS, and Android SDKs. Also available as raster layers.

Hail offering

Xweather Protect

Severe weather alerting and real-time storm monitoring for all your locations worldwide, with automated multichannel alerts and all-clear notifications through the Alert API add-on.

Learn more about Xweather Protect

Weather API

60+ endpoints of weather intelligence including hail threats, hail archive, lightning, severe storms, and road weather, delivered via a single RESTful API.

Learn more about Weather API

MapsGL

Interactive weather visualization for web, iOS, and Android. Includes eight dedicated hail map layers alongside hundreds of other weather and environmental layers.

Learn more about MapsGL

Integrate hail intelligence into your platform

Hail threat nowcasts, historical archive data, and map layers are available through the Weather API and MapsGL. Start building with a free weather API.